(Invited) Detailed Analysis of Highly-Efficient Emission in Organic Light-Emitting Diodes: The Dynamic Exciton Effect

Abstract

Thermally activated delayed fluorescence (TADF) is an excellent way to convert originally dark triplet excitons into light.1 To realize the conversion, minimization of the energy gap between S1 and T1, called ΔE ST, has been effective.1,2 Actually, reverse intersystem crossing (RISC) becomes possible by the small ΔE ST.1-4 However, the rate constant of RISC (k RISC) is relatively small compared to those of competing processes, such as radiative and non-radiative decays, resulting that RISC is the rate-limiting process in TADF systems. Therefore, the acceleration of RISC is a key factor to obtain further excellent TADF-based organic light-emitting diodes (OLEDs).The small k RISC originates from small spin-orbit coupling (SOC). The S1 and T1 of TADF molecules tend to have charge transfer (CT) character, because the HOMO and LUMO are well-separated to make the ΔE ST small. The intersystem crossing (ISC) and RISC between CT-type S1 (1CT) and CT-type T1 (3CT) are spin-forbidden.5 This problem can be solved by intervening locally excited (LE) state(s) between 1CT and 3CT. Recently, we successfully achieved very fast RISC with a k RISC exceeding 107 s-1 by a new material design concept, named “tilted face-to-face alignment with optimal distance” (tFFO)6-8 by realizing the excellent energy level matching of 1CT, 3CT, and 3LE with sufficient SOC between 1CT and 3LE.In the presentation, I will first talk about detailed analysis of a TADF molecule, MA-TA.8-10 The performance is very high; the experimentally-obtained photoluminescence quantum yield (PLQY) and maximum external quantum efficiency (EQE) were 100% and 23.9%, respectively. The results seem to be contradictory to the above, because donor and acceptor (D-A) segments are perpendicular with each other resulting that both the S1 and T1 are CT-type, and higher-lying 3LE states cannot participate in the RISC process. Also, highly PLQY of 100% cannot be expected from the perpendicular structure. Our detailed quantum chemical analysis revealed that the dynamic or static D-A fluctuation was the key trigger of both the fast RISC and high PLQY. Such direct 3CT→1CT RISC-based OLEDs without using any 3LEs have an advantage that various kinds of hosts having different polarities can be used without sacrificing the device performance, that is, high EQE and blue-shifted emission can be realized simultaneously.Secondly, I will talk about the development of a high-throughput material screening method by quantitatively predicting rate constants of all relevant electronic transitions. TADF molecules have been designed through the calculations of ΔE ST and oscillator strength to realize effective RISC and following radiative decay simultaneously. However, in addition to ΔE ST, SOC is also another key factor to control k RISC as described above. Not only radiative but also non-radiative decays should be considered to obtain high PLQYs. We here propose a new method to predict TADF performance more precisely.11 Our method based on the Fermi golden rule enables us to theoretically predict relevant rate constants for all types of electronic transitions in an emitter molecule with reasonable computational cost. We applied the method to benzophenone; the calculated rate constants quantitatively agreed with the experimental ones. We are now extending this method to various TADF molecules and singlet fission systems.12-14 We express sincere thanks to my group members. This work was supported by JSPS KAKENHI Grant Numbers JP20H05840 (Grant-in-Aid for Transformative Research Areas, “Dynamic Exciton”). Computation time and NMR measurements were supported by the international Joint Usage/Research Centre at the Institute for Chemical Research, Kyoto University, Japan.1 H. Uoyama, K. Goushi, K. Shizu, H. Nomura & C. Adachi Nature 492, 234 (2012).2 H. Kaji et al. Nat. Commun. 6, 8476 (2015).3 Z. Yang et al. Chem. Soc. Rev. 46, 915 (2017).4 M. Y. Wong & E. Zysman-Colman Adv. Mater. 29, 1605444 (2017).5 M. A. El-Sayed J. Chem. Phys. 38, 2834 (1963).6 Y. Wada, H. Nakagawa, S. Matsumoto, Y. Wakisaka & H. Kaji Nat. Photon. 14, 643 (2020).7 Y. Kusakabe, Y. Wada, H. Nakagawa, K. Shizu & H. Kaji Front. Chem. 8, 530 (2020).8 H. Imahori, Y. Kobori & H. Kaji Acc. Chem. Res. 2, 501 (2021).9 Y. Wada, Y. Wakisaka & H. Kaji ChemPhysChem 22, 625 (2021).10 Y. Wada, K. Shizu & H. Kaji J. Phys. Chem. A 125, 4534 (2021).11 K. Shizu & H. Kaji J. Phys. Chem. A 125, 9000 (2021).12 K. Shizu, C. Adachi & H. Kaji J. Phys. Chem. A 124, 3641 (2020).13 K. Shizu, C. Adachi & H. Kaji Bull. Chem. Soc. Jpn. 93, 1305 (2020).14 K. Shizu, C. Adachi & H. Kaji ACS Omega 6, 2638 (2021).